Since there are no immediate plans to expose generic transverse projections,
it’s slightly faster and more concise to inline this implementation.

Fixes #1157.

In some cases IE9 processes the opening of an XDomainRequest synchronously, so
by the time we were setting the response function to parse the DSV file, it was
too late. So instead we use the internal d3_xhr method to set the response
function before the request is sent. This is a continuation of fix #1260.

When computing the midpoint between two points both at longitude -π, there was a
chance that atan2 would return +π rather than -π. Now, we check whether the
points to interpolate lie (approximately) on a meridian; if they do, we can
approximate by using linear interpolation of longitude rather than atan2. This
fixes #1635 which was caused by new trig implementations in Chrome Canary.

Previously, invalid transform attributes were being set for an
ordinal→quantitative scale transition, because entering ticks were
undefined in the old (ordinal) scale, and exiting ticks were undefined
in the new (quantitative) scale.

Fixes #1620.

Rather than both setting the user-select to none and suppressing the selectstart
event, we now prefer to suppress the selectstart event if supported. This avoids
redundantly setting the user-select style, which can be slow. Fixes #1599.

Add unit test cases for % format.

Add a unit test case for it.

Rather than use d3_geo_transform, parts of d3.geo that need a point-based
geometry stream transform now use d3_geo_transformPoint, which is slightly more
efficient. (This is the same as the d3_geo_streamTransform that was removed in
ac9f51fa.)

When resampling is turned off, d3_geo_resample can return a special
transform that only performs projection of points to 2D, rather than
performing any additional computations such as conversion to 3D
coordinates for resampling.

We we’re distinguishing between the case where the cell completely surrounds the
clip extent and where the cell does not intersect the clip extent at all.
Related #1578.

Also, add a number of missing imports: mainly "math/abs".

Fixes #1581.

If data is updated so that some nodes no longer have children, and a
custom children accessor is used, then the children array was not being
removed for such nodes, which meant that hierarchical layouts were
incorrectly taking into account these children.

Fixes #1579.